arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2608.12438cs.LGhep-phstat.ML

用路径积分统一生成模型

Unifying Generative Models with Path Integrals

Ramon Winterhalder

首次发表
浏览论文内容

中文总结 AI 辅助

该研究将生成建模表述为路径积分,提出MSRJD形式的图示微扰理论,为确定性采样器提供无随机采样成本的单圈修正,还推导了响应加权得分匹配目标与EFT幂次计数下的对称等变漂移设计。

中文摘要 AI 辅助

我们将生成建模表述为一种路径积分,其中基于流、扩散、变分和对抗的模型,均源于对单一主作用量的不同评估原则。其Martin-Siggia-Rose-Janssen-de~Dominicis(MSRJD)形式将自由概率流与相互作用概率流分离,使其可纳入图示微扰理论。该展开在无随机采样成本的情况下,为确定性采样器提供了单圈修正,我们在可解与非线性漂移上验证了这一点,其将53%的树级误差降至1.6%。不完善的学习得分作为插入项,产生响应加权得分匹配目标,而对称等变漂移设计则成为有效场论(EFT)幂次计数下的算符展开。

英文摘要

We formulate generative modeling as a path integral in which flow-based, diffusion-based, variational, and adversarial models arise as different evaluation principles for a single master action. Its Martin-Siggia-Rose-Janssen-de~Dominicis (MSRJD) form separates free from interacting probability flows and opens them to diagrammatic perturbation theory. The expansion yields a one-loop correction to deterministic samplers at no stochastic-sampling cost, which we validate on solvable and nonlinear drifts, where it reduces a 53 % tree-level error to 1.6 %. Imperfect learned scores enter as insertions and yield a response-weighted score-matching objective, and symmetry-equivariant drift design becomes an operator expansion with EFT power counting.

发表机构

  • Università degli Studi di Milano(米兰大学)
  • INFN(意大利国家核物理研究院)

机构由 AI 辅助整理,请以论文原文为准。

补充信息

↑